Professional Documents
Culture Documents
Motivations and Justifications Behind A Concept Introduced by Gyorgy Ligeti PDF
Motivations and Justifications Behind A Concept Introduced by Gyorgy Ligeti PDF
Released by the author under the terms of the GNU General Public Licence
Contents
1 Polyphony 1
1.1 Polyphony and Perception . . . . . . . . . . . . . . . . . . . . 1
1.2 Two-Voice Polyphonies . . . . . . . . . . . . . . . . . . . . . . 1
1.3 More Voices . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.4 Chopin’s Atonal Movement . . . . . . . . . . . . . . . . . . . 3
1.5 How Fast Can We Play and Hear? . . . . . . . . . . . . . . . . 3
1.6 The Electronic Music Studio of the W.D.R. . . . . . . . . . . 5
1.7 “Fake” Polyphony . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.8 Closing Thoughts on Ligeti’s Experiences in the Electronic
Music Studio . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.9 Influences of the Electronic Music Procedures on Ligeti’s In-
strumental Works . . . . . . . . . . . . . . . . . . . . . . . . . 12
4 Conclusions 33
ii
5 Appendices 35
5.1 Granular Synthesis . . . . . . . . . . . . . . . . . . . . . . . . 35
5.2 Physical Modelling Synthesis . . . . . . . . . . . . . . . . . . . 36
5.3 Csound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.4 Stochastic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
iii
List of Figures
iv
3.7 Motif widely used by Ligeti in many of his pieces. . . . . . . . 21
3.8 String Quartett N.1, First Movement, bars:7-8 . . . . . . . . . 21
3.9 Motif B.A.C.H. very widely used in the music history . . . . . 22
3.10 Atmosphères Sample microstructure. Bars: 44-46 Violins I-1
to I-6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.11 Motion of the violins I and II (14 + 14, 28 solo parts). Bars:
44-45. A general descending motion. . . . . . . . . . . . . . . 23
3.12 Overview of bars 44-45. Strings section all solo 48 parts. Vi-
olins I and II are in descending motion while violas and cellos
are ascending. At the point where the graph ends the ascend-
ing motion of the violas and cellos parts will extend to the
violins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.13 A sketch drawing of the evolution from bar 44 to 49. It should
be noted that the widening of the boxes is to represent an
augmentation in the density of the parts . . . . . . . . . . . . 24
3.14 Bars: 44-47. Cluster notation shows that there is only a few
significant shiftings in ranges. . . . . . . . . . . . . . . . . . . 25
3.15 Some rhythmical Resultants according to the Schillinger Sys-
tem of Musical Composition[6] . . . . . . . . . . . . . . . . . . 27
3.16 String Quartett N.2, First Movement, bar: 43. Uneven subdi-
visions of the beat. . . . . . . . . . . . . . . . . . . . . . . . . 28
3.17 Sample rhythms of the Violins II and Viola parts . . . . . . . 29
3.18 Atmosphères Time-intensity view of the complete piece. . . . . 30
3.19 View of the section analyzed. After the stopping of the cluster
in the double-basses the 48-voice polyphony builds up a big
crescendo. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.20 Sonogram view of the micropolyphonic section. Darkened ar-
eas show a condensation in the high frequencies at those times
when several instruments are playing ponticello. . . . . . . . . 32
3.21 Spectral view of the same section. The jagged area also shows
the increase in ponticello playing. . . . . . . . . . . . . . . . . 32
v
Abstract
1
Requiem, Apparitions, Lux Æterna, (parts of) String Quartett N.2 and many others
2
Rehearsal letters H to J
3
Aleatoric or partly aleatoric notations
Chapter 1
Polyphony
X X X
§ 43 X X X X ü
yX 3 X X X
4 X X
Bach, in his few two-voice fugues seems to literally unleash each of the
1
two-parts but even in his three-voice polyphonic writings he often “links” two
of the parts in an homophonic evolution.
|
´ X ´
§ ´ 43
e
X ´
y ´´ 3 ´
4 Á
i
§ ´ 1 ii i
y 1
´
2
1.4 Chopin’s Atonal Movement
One striking example of mid-nineteenth century is the last movement of
Chopin’s Piano Sonata N.2 in B-flat Minor Op.28.
Ligeti calls this piece “. . . probably the first atonal piece of in the history
of music”1
Although there is a clear harmonic structure which appears in the quasi-
arpeggio evolution of the identical lines of both hands, the speed of harmonic
changes and their intently chromatic profiles actually makes the “feeling” of
tonality to weaken to a considerable degree.
The perception of such a music is more “global” rather than detailed
(structural). The popular epithet“winds blowing through graves. . . ” actually
gives a rather poetic yet accurate description of how this music was (is)
perceived.
Presto
´´
§´´´H
y ´ ´ H X
´ ´ ´ X
Figure 1.4: According to Ligeti music history’s first atonal piece: F. Chopin
Sonata N.2 in B-flat Minor, Fourth Movement “Presto”
Beside being called “the first atonal piece” by Ligeti, the Chopin example
above brings us to the very important question of sheer musical speed and
perception.
3
The seeing of a continuous action in cinematography is possible at a rate
of at least 20 images per second. The industry standard in motion picture
is 24 images per second. Going down from that speed we start to see first
a blurred and jerked motion and, down again, we start to distinguish each
actual frame.
Although it is closely related to instrumental specifics, note intervals,
hand positions and so on, a speed of twenty notes per second is not only
our playing speed-limit but also our limit for distinguishing individual mu-
sical events. Faster than this, the events blur in a glissando texture. This
particularity is often used in trills and glissandi.
= 60
20
§ 1
Figure 1.5: According to recent research the fastest notes we can distinguish
as individual elements
In the piano literature, some examples stretching the limits for the fastest
playing possible are worth mentioning here.
X X
X
y´6 X X X
¾ X X
X
å
8
y 6
´8 X
Figure 1.6: F. Chopin Prelude N.24 in D Minor. One of the fastest scales a
pianist has to play. . .
4
11
11
§ 8
6
y6 e
8
2 2
the last movement of the Petrouchka suite by Stravinsky we have, at the right
hand, eleven notes to be played at approximately one second time, the usual
tempo of the passage is 60 BPM. This requested speed is almost identical to
another passage, for the left hand, in the first movement of the same piece.
§ 42 Á
Á
y 2
4
5
fragments of magnetic tape which were then manipulated in any possible
way. Mixing and changing the playback speed were the most often performed
manipulations.
In 1957, Ligeti was assisting Gottfried Michael Koenig in the electronic
music studio of the Westdeutscher Rundfunk in Cologne for the realization
of a piece called Essay.
Essay by Koenig was mainly composed of sinusoidal sounds which can be
easily followed by the ear at some points but at other moments, those sounds
formed interesting conglomerates where none of them could be individually
spotted. That was due to the extreme brevity of them and to the high speed
of their succession in time.
In Essay, the simultaneity of sounds above and below the differentiation
level of the ear1 created an interesting texture.
In this texture one could spot and follow those sounds above the per-
ception threshold, sounds longer than 50ms., as a kind of “melody”. At the
same time, sounds below that threshold, shorter than 50ms., are perceived
as if they were “simultaneous” with the others. Even when they were not.
That “false simultaneity” not only created an interestingly complex sound
aggregate, but another quality, previously unheard of, emerged as well.
Koenig called this quality the “sound (timbre-color) of movement” Bewe-
gungsfarbe [1]. Rhythm is not anymore perceived as a motion but instead
rather a constantly changing sound-color is heard.
More layers of this composite “sound-color” could be created by varying
the dynamics of the component waves.
Seen from today, it seems that Koenig and Ligeti have “discovered” the
principle of Granular Sound Synthesis 2 .
In Granular Synthesis extremely brief sounds, when agglomerated create
a layer of a complex timbre.
Each of the “grains” can be varied in pitch, intensity and overtone com-
ponents3 while the overall pitch and intensity of the whole complex can also
be made continuously changing.
Today, Granular Synthesis is widely used tool. Various computer software
exists and it has various implementations in the Csound electronic music
software4
In an article formerly published in a compilation named Rückblick in die
Zukunft 5 , republished in[1], Ligeti describes the basic method of working in
1
Longer or shorter than 50 ms.
2
See Granular Synthesis, Appendices page:35 for more information
3
harmonic composition in the acoustical sense
4
See Csound, Appendices page:38 for more information.
5
Severin und Siedler, Berlin, 1981
6
the Cologne studio:
By using different intensity levels for each note and each sound,
and by integrating sounds which are harmonic, subharmonic and
non-harmonic in between a succession of sinus waves, it is pos-
sible to create a “fake” polyphony inside a monophony. This is
possible because of the mental connexions at higher levels that
we can create between different kinds of sounds.
1
See: Figure:1.9, page:8
2
overtone
7
If the higher level signals are not too dense we perceive, inside
the true single voice, a complex of several voices. By creating
a more detailed sequence we reach a saturation point where the
shape of the super-signals1 start to blur, too many layers neu-
tralise each other and the phenomenon of “fake” polyphony is
gradually lost.
8
Figure 1.10: Pièce électronique N.3. Fragment of the “score”.
The “score” for this Pièce électronique N.3 is drawn, as it was customary
at that time for electronic pieces, by using a squared millimetric paper. The
vertical axis represents the frequencies, the horizontal lines represent the
length of magnetic tape fragments in centimeters. Vertical lines are the
connecting points for the assembly.
The “score” is actually a matrix view 1 of the music.
While the majority of the sounds used are sinus waves, some electronically
filtered noises also appear. Those noises were filtered using a very narrow
band filter and they are actually very close to sinus waves but, as described
by the composer, they are made “slightly blurred by an halo of external
frequencies”.
The textures created are almost never static but always moving, evolving
and transforming.
The frequencies are hardly legible in the reproduced score2 which is shrank
here to fit the page but it is interesting to note that frequencies were selected
in order to produce various spectrums of harmonic and non-harmonic struc-
tures.
For harmonic structures, frequencies with plain integer ratios between
them are used, higher harmonics are privileged in this piece. Those harmonic
and non-harmonic spectræ create a resulting “sound” which is not “created”
as the others but can only be heard at the actual audition of the piece. This
is a “resultant” harmonic sound.
In autumn 1957, the“score”for frequencies were established and in Novem-
ber Ligeti started the realization of the piece in the Cologne studio.
1
Term used in computer “sequencer” software together with the name “piano roll view”
to show and edit each note in a pitch-time display
2
Figure: 1.10
9
He was first creating the sine waves at the preset frequencies without
dealing with the intensities. But, soon the technology of the time made it
impossible to continue.
The synchronizing of more than forty layers of sound was a practical im-
possibility with the technology of the fifties. The out-of-sync shifting created
audible differences in the resulting textures which were deemed unacceptable
by the composer.
After having completed the work on another electronic piece: Artikulation
in 1958, Ligeti resumed working on Pièce N.3 .
This time he confronted with another difficulty: the scoring of the intensi-
ties were unrealizable because that necessitated working in three-dimensions. . .
On the surface of the quadrilled paper, pitches were assigned to the vertical
axis and time (durations) to the horizontal one.
So he started to notate the intensities on another sheet in close connex-
ion with the frequency-time graph. But he stopped working altogether on
that music and also abandoned completely the electro-acoustic realm. The
practical impossibilities of composing music exactly as he wanted to do com-
bined with the technical limitations of the equipment made him leave the
electro-acoustic music studios forever.
He relates his leaving the studio in the article mentioned before[1]
As other tentatives for the realization of the piece revealed to be
unsatisfactory as well, I promptly abandoned working on it and
from summer 1958 I totally concentrated on the composition of
pieces for orchestra.
10
It is interesting to denote that while many of the composers with Ligeti
among them, were attracted to the electronic music studios because there
were a possibility for them to accurately create their music as exactly as
they wanted it to be, without the intervening of the performer to “sonify” it,
Ligeti did not accept to remodel his musical conceptions to fit them in the
technological realities of that time.
In this aspect, there is a strong connexion between him and Pierre Boulez
in their reasons to depart form the electronic music studios.
More specifically, Ligeti, in his Réflections faites au départ du studio et
plus tard [1] first mentions the impossibility to work with (sound) envelopes
and attack transients.
Having subtle timbre modifications was also a practical impossibility.
The sheer quantity of individual single sine-waves which has to be created,
recorded on tapes and mixed for making a complex timbre and its modifica-
tions in time was simply not feasible at that “pre-computer” time.
He also states another limiting fact which remained with most the elec-
tronic music instruments for a very long time. Before the development of
Frequency Modulation first and spectral manipulation later, the electroni-
cally created sound was, in essence, a static one.
Commonly used voltage controlling devices for pitch (VCO), intensity
(VCA) and timbre (VCF) were almost crude and too regular in their periodic
oscillations.
Frequency Modulation (FM) technology allowed for the creation of the
first electronic timbre which had the capabilities of controlled aperiodic os-
cillations evolving in a aperiodic manner.
The famous bell sounds of the first commercially available portable syn-
thesizer featuring the FM sound synthesis, Yamaha DX7TM , were world fa-
mous for a reason. They were the first synthetic sounds of the kind.
However the wide timbral capabilities Ligeti was after have been exten-
sively and widely available only with the advent of Physical Modelling 1 , Gran-
ular Synthesis 2 and all the spectral (re)modeling tools we have today. They
only flourished starting from the 90’s.
Ligeti metions the book The Technology of Computer Music by Max V.
Mathews3 and the software Music-V as forerunners of a new area in electronic
music composition.
The Music-V system was to give birth to Csound 4 .
1
See: Physical Modelling Synthesis page:36 for more information
2
See: Granular Synthesis page:35 for more information
3
Cambridge, London, The M.I.T. Press, 1969
4
See: Csound page:37 for more information
11
1.9 Influences of the Electronic Music Proce-
dures on Ligeti’s Instrumental Works
In his essay Musique et Technique, originally published in a compilation
called Rückblick in die Zukunft 1 Ligeti relates that when two of his orchestral
compositions Atmosphères and Apparitions were first performed, in 1961
and 1960, several listeners thought there were loudspeakers disseminated in
between the performers.
The illusion was caused by the composer using techniques he already
worked on in the Electronic Music Studio of Cologne, techniques he succes-
fully adapted to instrumental music.
He calls that specific technique the “timbre of motion”.
[. . . ] by getting below the threshold of “fusion”2 we obtain modi-
fications of the structure of the musical texture and we get instru-
mental combinations which are sensibly different from those ob-
tained by using the conventional instrumental combination tech-
niques.
In the mentionned article, he first uses the term micropolyphony:
Already from 1950 I was imagining new timbres, a“static”musique,
“sound spaces”3
In Budapest, 1956, before my electronic music studio experiences
I composed a piece called Viziók (Visions) which later became
the first movement of Apparitions.
[. . . ]
I applied those experiences to orchestral music, in 1959 in a re-
ally different way to the second movement of Apparitions right
after my stay in Cologne. In the first movement the sound spaces
were still static and aligned as successive blocks while, in the
second movement, for the first time, the destabilization of those
blocks appeared as well as textures modifications and the microp-
olyphony.
[. . . ]
The experiences made in the studio of electronic music by using
the “melting of succession”4 and by stacking a great number of
1
Severin und Siedler, Berlin, 1981
2
Distinctly perceptible sound
3
The term will then be widely used in its abbreviated form: soundscapes.
4
fusion de successivité
12
independently elaborated sounds and sound sequences on top of
each other, made me imagine a kind of complex polyphony cre-
ated by musical textures and networks. I called this way of com-
posing micropolyphony because the many rhythmical elements
were getting below the threshold of fusion in the polyphonic tex-
ture.
The texture gets so dense that the parts can not be perceptible
anymore in their own individuality and we can apprehend the
whole only as a “whole”, from a higher perception level.
13
Chapter 2
Micropolyphony: “Microscopic”
Polyphony
1
See:Xenakis Sound Clouds and Ligeti’s Micropolyphony 3.3 page:18
14
2.2 Precursory Attempts
Ligeti mentions an interesting orchestral effect, one from the last periods
of Romanticism in music: Feuerzauber at the end of the Walkyrie by Wagner.
He describes the dazzling effect as:
15
Chapter 3
16
parameters are evolving and can be continuously changing within the shape
as well. So there is no real way to represent it graphically.
This aspect of continual changes and evolution is what makes the Ligeti
soundscapes first and unique in the history.
1
Notes partly defined in pitch rhythm or succession
17
3.3 Xenakis’ Sound Clouds and Ligeti’s Mi-
cropolyphony Compared
3.3.1 Stochastic Methods
Another way of creating a similar soundscape was devised by Iannis Xe-
nakis using his mathematician and architect background.
To create similarly dense structures, Xenakis used stochastic processes
and formulæ1 .
Stochastic functions are basically conditional random transition tables.
The basic rule for them is that an element ε may be followed by another
element ω with a probability factor ρ. This is commonly expressed as: x ⇒ y ρ
Those tables of probabilities are better known as Markov Chain Processes.
From To Probability
a→ b⇒ .25
d→ c⇒ .30
e→ f⇒ .4
x→ z⇒ .05
a→ f⇒ 0
Figure 3.2: Example of a simple Markov chain transitions table. The first
line read as: “there is a 0.25 chance (25%) that the element a
changes to (or will be followed by the) element b”
1
See: Stochastic, Appendices page:40 for more information
2
Specially in the Music of Changes
18
Figure 3.3: I. Xenakis Drawings for the strings glissandi in Pithoprakta
19
22
W
X
3 5
´ ´
´
§ Á Á
5 5 5
´ 3 ´
X X X
§ Á Á ´ ´
3 5
Á Á
5 5 3
Figure 3.5: G. Ligeti Atmosphères, Violin I-1, Bar:44. Despite being only
one component of the 48-voice polyphony, this line of the number
one of the first violins still shows a clear melodic structure.
playing a mirror canon of the violin parts. When this cello section is examined
on its own, it reveals the coherent micro-structure of the polyphony.
legatissimo
W (con sord.)
K 2 ´
3 3
2
Á
K ´
X X
K
X
X
Á
X
K W X
X
Á
K X W X
X
Á
Á
´
K X W
X
e
e
e
´
§ X
´
´
§ ´
´
e
´
´
b We
§
e W
´ ´ b
§
However one may ask the question “is this audible and if so in what
sense?”. It is obivous that the structure shown, one part of the forty-eight
voice polyphony, is literallt buried and would never be recognized in the mass
of sound.
20
The point in crafting each part of the micropolyphony to such an extend is
to control the shape and evolution of the “soundscape”. Individual notes and
parts, altough not discernable in their own, are creating the characteristics
of the “colour”, “shape” and “inner activity” of the soundscape.
Range, rhythm, dynamics and intervals for each part are not meant to be
heard as such but rather to be a part of the overall design.
3.4 Microstructures
3.4.1 Intervals
One familiar “Ligeti motif” immediately captures attention. The motif:
X
§ X
This “favorite” motif by Ligeti is the main structural element of his First
String Quartett (Métamorphoses nocturnes 1953-54). It is the pc-set (0, 1, 2, 3),
Forte Code: 4-1, interval vector: [321000]; a highly versatile set.
X X
§ 43
p dolce
§ 43
X X X X
K 3 X
4 X X
K 3 X X
4
21
§ ´
Figure 3.9: Motif B.A.C.H. very widely used in the music history
X ´ ´ ´
Vl.I-1 §
´ ´ ´
Vl.I-2 §
´ ´ ´
Vl.I-3 §
´ ´ ´
Vl.I-4 §
´ ´ ´ ´ X
Vl.I-5 §
´ ´ ´ ´
X
Vl.I-6 §
(rhythmically condensed form of Viol.1-5)
The intervals of the prime form of the set (whole-tone, semi-tone, whole-
tone) are used in the form of permutations throughout the section.
It is interesting to note that the whole section (bars: 44-48; up to the
re-entrance of the Double Basses at bar: 49, rehearsal letter: I) is composed
exclusively with neighbor tones, half and whole.
The few “jumps” which occur at bars 48 (cello 10 Violin II-14) or 49
(Violin I-14) are clearly for shifting the line back into the range of the in-
strument. The shift at the second half of the bar 44, Violin II-12-13 and 14
are all marked imperceptible attack on the score.
At the beginning of the section; bar: 44-45, violins I and II are moving
together in a generally descending motion while violas and cellos are set in
an upward motion
The upward motion of the viola and cello parts creates a “rising wave
motion” which gradually expands to the violin parts starting from bar 47.
The graph, figure:3.13 (page: 24) shows the overall evolution of the sound-
22
Figure 3.11: Motion of the violins I and II (14 + 14, 28 solo parts). Bars:
44-45. A general descending motion.
Figure 3.12: Overview of bars 44-45. Strings section all solo 48 parts. Violins
I and II are in descending motion while violas and cellos are
ascending. At the point where the graph ends the ascending
motion of the violas and cellos parts will extend to the violins.
cloud.
23
Figure 3.13: A sketch drawing of the evolution from bar 44 to 49. It should
be noted that the widening of the boxes is to represent an aug-
mentation in the density of the parts
24
Bar: 44 Bar: 45 Bar: 46 Bar: 47
Vl-I. §
Vl.II §
Vla. K
Vc.
K
y
CB.
Figure 3.14: Bars: 44-47. Cluster notation shows that there is only a few
significant shiftings in ranges.
25
The book has many insights on the process of artistic creation, the history
of music and arts.
Few sample citations will enlighten this very important music thinker’s
approach by doing so it will also help to set Ligeti’s attitude and some of his
compositional techniques.
Joseph Schillinger wrote in his The Mathematical Basis of the Arts[7]
The history of art may thereupon be described in the following
form:
1. Nature produces physical phenomena, which reveals an aes-
thetic harmony to us [. . . ] This is the pre-aesthetic, natural
(physical, chemical, biological) period of art creation.
2. Man recreates aesthetic realities by reproducing the appear-
ance of the physical realities through his own body, or through
a material at his command [. . . ] he expresses the laws of
mathematical logic through his sensory experience; this is
the intuitive period of art creation.
3. Becoming more and more conscious in the course of his evo-
lution, man begins to create directly from principles [. . . ]
This is the rational and functional period of art creation
[. . . ]
It is time to admit that aesthetic theories have failed in the
analysis as well as synthesis of art.
[. . . ]
The cult of craftsmanship transforms into formalism and scholas-
ticism.
Despite his outstanding contributions Schillinger is curiously unknown
in music theory circles. Among those contributions are: a very systematic
and complete theory of voice-leading by using simple circular permutation
techniques and modulo arithmetic, an interesting theory of pitch scales and
symmetrical divisions of one or more octaves1 , interesting investigations on
the semantics of the music, enlightening opinions on orchestration and in-
strumentation. . .
However his most important contributions are probably in the field of the
musical rhythm.
Rhythmic Resultants
A new and interesting idea of Schillinger is the rhythmical resultant[6].
1
Symmetrical scales theory have been widely spread by others but Schillinger, if not
the first was the one who systematized it to a much greater extend
26
Two different rhythms, when heard at the same time, create another
which is the resultant of the two.
Rhythm:3
Rhythm: 1 Õ
i
Rhythm:4
i i
Rhythm:2 Rhythm:3
Rhythm: 2 Õ
Resultant Õ
Resultant: 2-3 Resultant: 4-3
Ligeti, in many of his works have used the resultant rhythms techniques in
very practical ways. In his Second String Quartet, for instance, the uneven
subdivisions of the beat serves to create an asynchronous yet still “calm”,
animation.
In the fragment shown below1 and in the following bars, we often get a
stacking of subdivisions 4/5/6/7 of the beat.
That makes no single attack, except one occurring on the beat, of any part
simultaneous with any other. This is a very effective and practical way to
create an illusion of a faster playing speed. In such a beat, when players play
simultaneously four, five, six and seven notes on a beat, the actual rhythm
which is perceived is actually 4 + 5 + 6 + 7 = 22 minus 3 (for the “on-beat”
simultaneous attack) equals 19 distinct notes in a beat.
If the metronome speed is one second a beat or faster, the resulting“speed”
of the notes, 19 notes per second would almost equal the perception threshold
which is of 20 discernible “notes” per second, or 50 millisecond for a musical
event.
As stated earlier2 experiments made in the electro-acoustic music studios
showed that for durations below 50ms. musical elements, in our perception
tend to dissolve into a continuum.
This technique, being a highly effective and “economical” one, because it
surpasses performers and instruments physical limitations for the creation of
an over-fast animation, is extensively used by Ligeti. Most notable areas of
its use are for creating over-fast pizzicato sections which would be otherwise
impossible to play, over-fast trills and huge wave-like motions.
1
Figure:3.16, page:28
2
See: section 1.5, page: 3.
27
W
molto legato, molto calmo, sempre senza vibr.
ü ü ü
5
§ X
5
X X X
W
ppp
§ å 5
´ X
6
ü ´ ´ ´
X X
ppp
K X
7 6 7
´ ´ X X ´ ´
X X X
6
´ ´
y X X X X ´ ´ ´ ´ ´ ´
X
6 6
7 7
Figure 3.16: String Quartett N.2, First Movement, bar: 43. Uneven subdi-
visions of the beat.
28
Õ
ÁÁÁ 3
3 3 4 3
3
more quintuplets of 32nd. notes appear in the now even denser polyphony.
Double-basses are set in motion, yet not so fast due to the inertia of the
instrument at the last two bars of the section.
The crescendo at bars: 52-53 is made stronger by the “shrinking” of the
“cloud”. As more individual instruments are joining on the same notes, near
the end of bar:53 all instruments conglomerate in a very narrow range around
middle-C.
The texture style created with the rhythms shown above, in the second
Violins and Violas, is best described as an “ornamental style”. Those rhythms
combined with the neighboring intervals shown in the section: 3.4.1, page:
21 give the “cloud” its characteristic “inside-motion” effect.
The crowded appearance of the notated rhythm is intended to get even
more “busy” sounding at the performance for each instrumentalist will in-
evitably slide from the exact notated values. Syncopes and tuplets will make
for them impossible to be on the exact rhythm.
On the other hand, the detailed scoring will also prevent the passage to
“deflate” by having musicians playing in synchronization with their neighbors.
29
Figure 3.18: Atmosphères Time-intensity view of the complete piece.
Playing sul tasto gives the section its distinctive color. Then, at bars: 49-
50; violins I, II and the cellos start getting one by one (beginning from the
second violins) to play sul ponticello. A gradual shift of sound-color occurs.
This shift is notated as a very detailed process. It starts at the second
violins then it “spreads” to the cello section and almost at the same time,
starting from the Violin-I N.14 and going up to Violin-I N.1 it creates a
gradual change in sound-color, as if the acoustical output from the orchestra
was miked (recorded) and then modified, filtered through an electronic device.
The process is wanted to be really continuous, so when Violins I are
getting “more and more” on ponticello, at the same time, cellos are getting
back to normal playing position. This is noted poco a poco ord. in the score
(bar: 49).
This color-shifting is also emphasized by the entrance of the Double Bass
section in a cluster-like chord sul tasto and pppp at bar: 49.
From this time on, ponticello and normal playing is alternatively dis-
tributed among sections.
This creates a rather unique sound-effect which is best described by Ligeti
himself:[1]
30
Figure 3.19: View of the section analyzed. After the stopping of the cluster
in the double-basses the 48-voice polyphony builds up a big
crescendo.
31
Figure 3.20: Sonogram view of the micropolyphonic section. Darkened areas
show a condensation in the high frequencies at those times when
several instruments are playing ponticello.
Figure 3.21: Spectral view of the same section. The jagged area also shows
the increase in ponticello playing.
32
Chapter 4
Conclusions
33
Micropolyphony is only one of the techniques experimented on or discov-
ered in the electronic-music studio and then “adapted” to orchestral writing.
In many of his choral pieces, for instance Lux Æterna or in the Requiem
the way he handled part entrances and exits is yet another such application.
Parts are fading in and out as if manipulated by a mixing desk. Gradual
shifts of tone-color is yet another adaptation of the electronic music studio
techniques. This is an application of the band-pass/reject filters.
34
Chapter 5
Appendices
35
Some Granular Synthesis Software
• REplay PLAYer generative granular synthesis software for Mac
• Granulab real-time granular synthesizer for Win32
• Chaosynth cellular automata granular synthesizer by Eduardo Reck
Miranda.
• Vocal Modeler Special vocal effect for Reaktor that uses granular syn-
thesis.
• crusherX-Live! granular synthesis system for Windows
• CDP granular synthesizer from the Composer’s Desktop Project
• WSOLA time scale modification of audio using granular synthesis
• White FX a granular effect for Reaktor
• Audiomulch a real-time audio processing tool which has some so called
contraptions which offer granular synthesis.
• Atomic Cloud is an easy to use real-time grain cloud generator for
Windows
• Cecilia provides one of the best frontends available for employing gran-
ular synthesis. It uses the CSound language.
• Reason, from Propellerhead Software released a virtual device called the
Malström. The device, dubbed a “Graintable” synthesizer, combines
granular and wavetable synthesis technologies.
36
Thereafter the properties of the membrane (mass density, stiffness, etc.), its
coupling with the resonance of the cylindrical body of the drum, and the
conditions at its boundaries (a rigid termination to the drum’s body) would
describe its movement over time and thus its generation of sound.
Similar stages to be modelled can be found in instruments such as a
violin, though the energy excitation in this case is provided by the slip-stick
behavior of the bow against the string, the width of the bow, the resonance
and damping behavior of the strings, the transfer of string vibrations through
the bridge, and finally, the resonance of the soundboard in response to those
vibrations.
Although physical modelling was not a new concept in acoustics and syn-
thesis, having been implemented using finite difference approximations of the
wave equation by Hiller and Ruiz in 1971, it was not until the development of
the Karplus-Strong algorithm, the subsequent refinement and generalization
of the algorithm into the extremely efficient digital waveguide synthesis by
Julius O. Smith III and others, and the increase in DSP power in the late
1980s that commercial implementations became feasible.
Yamaha signed a contract with Stanford University in 1989 to jointly
develop digital waveguide synthesis, and as such most patents related to the
technology are owned by Stanford or Yamaha.
The first commercially available physical modelling synthesizer made us-
ing waveguide synthesis was the Yamaha VL1 in 1994.
While the efficiency of digital waveguide synthesis made physical mod-
elling feasible on common DSP hardware and native processors, the con-
vincing emulation of physical instruments often requires the introduction of
non-linear elements, scattering junctions, etc. In these cases, digital waveg-
uides are often combined with FDTD, finite element or wave digital filter
methods, increasing the computational demands of the model.
Examples of physical modelling synthesis:
3. Formant synthesis
5.3 Csound
Text copied from Wikipedia complying with the GNU General Documentation
License[2]
37
Csound is a computer programming language for dealing with sound, also
known as a sound compiler or an audio programming language. It is called
Csound because it is written in the C programming language, as opposed to
some of its predecessors.
Csound was originally written at MIT by Barry Vercoe, based on an
earlier language called Music360, developed by Max Mathews at Bell Labs.
It is free software, available under the LGPL. Its development continued
throughout the 1990s and 2000s, led by John ffitch at the University of Bath,
resulting in the launch of Csound 5 in February, 2005.
Many developers have contributed to it, most notably Istvan Varga, Gabriel
Maldonado (who developed a variant of the system, CsoundAV, which in-
cludes image and graphics processing extras), Robin Whittle, Richard Karpen,
Michael Gogins, Matt Ingalls, Steven Yi and Victor Lazzarini.
Csound takes two specially formatted text files as input. The orchestra
describes the nature of the instruments and the score describes notes and
other parameters along a timeline. Csound processes the instructions in
these files and renders an audio file or real-time audio stream as output.
grain3
grain3 Generate granular synthesis textures with more user control.
Description: Generate granular synthesis textures. grain2 is simpler to
use but grain3 offers more control.
Syntax: ares grain3 kcps, kphs, kfmd, kpmd, kgdur, kdens, imax-
ovr, kfn, iwfn, kfrpow, kprpow [, iseed] [, imode]
Initialization:
38
iseed (optional, default=0) seed value for random number generator (posi-
tive integer in the range 1 to 2147483646 (231 - 2)). Zero or negative
value seeds from current time (this is also the default).
Performance:
kphs grain phase. This is the location in the grain waveform table, expressed
as a fraction (between 0 to 1) of the table length.
kgdur grain duration in seconds. kgdur also controls the duration of already
active grains (actually the speed at which the window function is read).
This behavior does not depend on the imode flags.
39
kfrpow this value controls the distribution of grain frequency variation.
If kfrpow is positive, the random distribution (x is in the range -
1 to 1) is abs(x)((1/kf rpow)−1) for negative kfrpow values, it is (1 −
abs(x))((−1/kf rpow)−1) Setting kfrpow to -1, 0, or 1 will result in uni-
form distribution (this is also faster to calculate). The image below
shows some examples for kfrpow. The default value of kfrpow is 0.
kfn function table containing grain waveform. Table number can be changed
at k-rate (this is useful to select from a set of band-limited tables gen-
erated by GEN30, to avoid aliasing).
5.4 Stochastic
Text copied from Wikipedia complying with the GNU General Documentation
License[2] Accessed: February 15, 2008
40
random values of a stochastic process at different times may be indepen-
dent random variables, in most commonly considered situations they exhibit
complicated statistical correlations.
Familiar examples of processes modeled as stochastic time series include
stock market and exchange rate fluctuations, signals such as speech, audio
and video, medical data such as a patient’s EKG, EEG, blood pressure or
temperature, and random movement such as Brownian motion or random
walks. Examples of random fields include static images, random terrain
(landscapes), or composition variations of an inhomogeneous material.
41
References
[1] Ligeti, György. 2001. Neuf Éssais Sur La Musique. Genève: Éditions
Contrechamps.
[2] http://en.wikipedia.org.
[4] Vercoe, Barry, John ffitch, Jean Piché, Peter Nix, Richard Boulanger,
Rasmus Ekman et al. 1986, 1992. Massachusetts Institute of Technology.
[7] Schillinger, Joseph. 1976. The Mathematical Basis of the Arts. New York:
Da Capo Press.
42
Contact:
Mehmet Okonşar
pianist-composer
Mesnevi Sok. 46/15 TR 06690 - Ankara Türkiye
Tel. + 90 (312) 438 09 17 - GSM. + 90 (533) 767 18 99
www.okonsar.com
mehmet@okonsar.com